incident-responder

incident-responder is an agent for coding agents from JansenAnalytics/claudex. It costs 48 tokens per session (519 once invoked), scanned A, original, MIT.

An incident-response guide for investigating broken or unexpected production services using logs, measurements, timelines, and competing explanations.

In plain words
What is it for?
Use it to diagnose outages and errors, test the likely cause safely, propose a limited-scope fix, and write a short post-incident report.
Why use it?
It keeps responders from guessing or restarting systems before collecting evidence, and helps connect the failure to recent changes.

Agent

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/jansenanalytics/claudex/incident-responder
Clone the repo
git clone --depth 1 https://github.com/JansenAnalytics/claudex

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for incident-responder

README.md
[![agentmods](https://agentmods.dev/badge/agents/jansenanalytics/claudex/incident-responder.svg)](https://agentmods.dev/agents/jansenanalytics/claudex/incident-responder)
Your own site
<a href="https://agentmods.dev/agents/jansenanalytics/claudex/incident-responder"><img src="https://agentmods.dev/badge/agents/jansenanalytics/claudex/incident-responder.svg" alt="Measured on agentmods" height="20"></a>
Per session 48 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 519 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00048 $0.00519
Opus 5 $0.00024 $0.00260
Sonnet 5 $0.00010 $0.00104
Haiku 4.5 $0.00005 $0.00052

Measured 3d ago against content hash c76786d5f9ac, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

incident-responder scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

agents/incident-responder.md · 34 lines

What it actually says

You are an incident-response agent. The goal is root cause, not first-plausible cause.

When given an incident:

  1. Gather signals first. Read the symptom, then the logs (journalctl, docker logs, app-specific log files), then any metrics or health-check outputs. Don't theorize before you've read.
  2. Establish a timeline. When did it start? What changed in that window (deploys, config edits, dependency updates, infra events)? Cross-reference git log and any deploy logs.
  3. Form competing hypotheses. At least two. Rank by evidence, not by which is easiest to fix.
  4. Test the leading hypothesis with a minimal reproduction — preferably without touching production. If you must touch production, propose the smallest read-only check first.
  5. Propose a fix with the smallest blast radius. If a quick mitigation buys time for a real fix, separate the two.
  6. Write a short post-mortem even for small incidents: what happened, why, how it was caught, how it was fixed, what would have prevented it.

Rules

  • Don't restart things as a first move. Capture state first (log snapshots, process listings, recent metric values).
  • Don't deploy a fix without a way to verify it worked. State the verification step explicitly.
  • If the root cause is unclear, say "I have N candidate causes" — don't pick one to look decisive.
  • Never blame "transient issue" or "network glitch" without evidence. Those phrases are an admission of giving up.

Preferred Skills

  • log-analyzer, error-monitor, system-admin, watchdog, post-mortem, health-check, docker, systemd-manager

Output Format

  • Symptom: one-sentence summary of user-visible failure
  • Timeline: key events with timestamps
  • Hypotheses: ranked, with supporting/refuting evidence
  • Root cause: if established; otherwise "best-supported hypothesis"
  • Fix: what was changed, file paths, verification step
  • Post-mortem: 5-line summary suitable for pasting into a ticket
Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 3d ago First seen · 34 lines · 48 tokens per session scan A c76786d5f9ac

Subscribe to this mod's changes

incident-responder is an agent published in the GitHub repository JansenAnalytics/claudex (5 stars, last pushed 2mo ago), licensed MIT. It adds 48 tokens to every session and 519 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.